Tailored Reactivity of Ni+Al Nanocomposites: Microstructural

Sep 5, 2012 - Combustion of mechanically activated Ni/Al reactive composites with microstructural refinement tailored using two-step milling. B. Aaron...
0 downloads 0 Views 4MB Size
Article pubs.acs.org/JPCC

Tailored Reactivity of Ni+Al Nanocomposites: Microstructural Correlations Khachatur V. Manukyan,*,† B. Aaron Mason,‡ Lori J. Groven,‡ Ya-Cheng Lin,† Mathew Cherukara,§ Steven F. Son,‡ Alejandro Strachan,§ and Alexander S. Mukasyan† †

Department of Chemical and Biomolecular Engineering, University of Notre Dame, Notre Dame, Indiana 46556, United States School of Mechanical Engineering and §School of Materials Engineering, Purdue University, West Lafayette, Indiana 47907, United States



ABSTRACT: An efficient approach that combines short-term (minutes) highenergy dry ball milling and wet grinding to tailor the nano- and microstructure of Ni +Al composite reactive particles is reported. Varying the ball-milling conditions allows control of the volume fraction of two distinct milling-induced microstructures, that is, coarse and nanolaminated. It is found that increasing the fraction of nanolaminated structure present in the composite particles leads to a decrease in their ignition temperature (Tig) from 700 and 500 K. Material with nanolaminated microstructure is also found to be more sensitive to impact ignition when compared with particles with a coarse microstructure. It is shown that kinetic energy (Wcr) thresholds for impact ignition, obtained for an optimized nanolaminated microstructure, is only 100 J. High-speed imaging showed that the impact-induced ignition occurs through formation of hot spots caused by impact. Molecular dynamic simulations of a model system suggest that impact-induced localized plastic deformation raises the local temperatures to ∼600 K, enough to initiate exothermic reactions. Analysis of the kinetics and reaction mechanism shows that the reason for low Tig and Wcr for nanolaminated microstructure is the rapid solid-state dissolution of nickel in aluminum lattices.

1. INTRODUCTION Gasless (or low gas) high-energy density materials hold great promise for many applications, including chemical energy storage, microscale energetic devices, heat sources for joining of refractory substances, and synthesis of advanced materials.1−8 This class of energetic materials can exhibit extraordinary properties such as fast burning rates and high combustion temperatures, accompanied by low volume expansion or even densification due to the lack of gaseous products. Aluminumbased metal−metal reactive systems are of particular interest due to their highly exothermic nature associated with intermetallic formation.9,10 The combination of their mechanical properties and the ability to undergo rapid combustion reaction has prompted extensive studies aimed at investigating their use for design of structural energetic materials. One promising candidate for such applications is engineered reactive mixtures of nickel (Ni) and aluminum (Al).10−15 The exothermic reaction in the Ni+Al system may be initiated by a heat source14−16 or mechanical impact.1,17−19 In the first case, the ignition temperature is often considered to be a primary indicator of reaction sensitivity. The behavior of exothermic systems during external heating can be described by thermal explosion (or volumetric self-ignition) theory developed for gas-phase reactions20−22 and also adapted to the homogeneous condensed systems.23 These classical theories introduce several terms and conditions. For example, the difference between bulk thermal explosion and local ignition © 2012 American Chemical Society

phenomena needs to be delineated. Thermal explosion occurs when the thermal relaxation time of the reactive media is much smaller than the chemical reaction time. As a result, the entire sample ignites simultaneously. Otherwise, the reaction can be initiated locally; that is, the reaction begins at discrete locations within the sample that meet the ignition criteria. The reaction may then self-propagate throughout the sample in the form of a combustion wave with a defined speed.24 Most experimental studies have shown that the self-ignition temperature of micrometer-scale Ni+Al mixtures is equal to or above the eutectic temperature of the system (∼912 K).3,4,25,26 Therefore, the ignition phenomenon of the micrometer-scale mixtures is correlated to the formation of a liquid phase. Early experimental studies show that nickel rapidly dissolves into aluminum melt, and aluminum-rich Al 3 Ni and Al 3 Ni 2 intermediate phases were observed in quenched samples in the initial stage of the process.26 Before reaching the maximum reaction temperature (1911 K), the precipitation of solid AlNi phase from the supersaturated solution and the melting of the remaining solid nickel take place.3,26 Atomistic simulations also demonstrate the importance of Al melting in thermal and dynamically initiated composites.27−29 Received: April 10, 2012 Revised: July 11, 2012 Published: September 5, 2012 21027

dx.doi.org/10.1021/jp303407e | J. Phys. Chem. C 2012, 116, 21027−21038

The Journal of Physical Chemistry C

Article

during milling. Each batch was milled in one to four intervals consisting of up to 5 min of milling, followed by a 15 min rest to prevent excessive heating of the material during milling. Because a process control agent was not present in this stage, resulting particles were relatively large in size, having diameters up to 3 mm. After DM, the cooled jar was opened and 20 mL of hexane (98.5% hexane isomers, Mallinckrodt Chemicals) was added. Subsequent wet grinding (WG) in hexane was performed for a total of 10 min (two intervals of 5 min of milling, followed by a 15 min rest) with the same milling conditions, except that the jar was pressurized with argon to ∼0.3 MPa. One batch of the Ni+Al mixture was wet ground without preliminary DM for comparison. After such treatment, the samples were dried in air for 1 h and then vacuum-dried at ∼300 K, for 1 h to remove the hexane fully. As-milled powders were sieved into six fractions (Fi) with the following ranges of particle sizes (d): d < 25 μm (F1); 25−53 μm (F2); 53−106 μm (F3); 106−355 μm (F4); 355−850 μm (F5); and d > 850 μm (F6). 2.2. Reaction Initiation. The ignition temperature of the materials produced with high-energy ball milling was measured using a simple setup (Figure 1a) that includes a hot plate, a

In the case of mechanically induced reaction, the impact energy or projectile velocity may be used as a measure for the chemical reaction sensitivity. However, little is known about the effect of constituent properties on the threshold condition for reaction ignition. It was shown that for microscaled Ni+Al mixtures the threshold of impact velocities above which reaction may occur are greater than 1000 m/s (∼10 kJ).19 However, for successful design of structural energetic materials the threshold of impact velocities (energies) of Ni+Al mixtures must be decreased significantly. Recently, it was demonstrated that high-energy ball milling could enhance the reactivity and ignition sensitivity of the Ni +Al system.1,9,13,14,17 Such mechanical treatment significantly decreases both the ignition temperature (down to 600 K),14 as well as diminishes the value of critical impact energies (Wcr) required to initiate the reaction to 420 J.17 High-energy ball milling is a complex process that involves intensive interaction between the particles and milling media.30 It was shown that for mixtures with ductile compounds (e.g., Ni+Al powder mixture) flattening and cold welding of metal particles occurs in the initial stage of DM.30−32 An increase in composite particle sizes may also be observed in this stage. Longer processing causes strength hardening of the material. Consequently, its brittleness may increase, resulting in fragmentation of the composite particles. Further milling may lead to the formation of solid solutions, intermetallic, or amorphous phases or even cause sudden initiation of the exothermic reaction. 33 However, the influence of microstructural transformations on thermal and mechanical sensitivity of formed composite particles has not yet been fully investigated. Other open questions are how solidstate mass transfer alone can be responsible for the selfsustained reactions ignition in such systems and why they can be ignited by surprisingly low-impact energies. This work is a continuation of our efforts to more fully answer some fundamental issues with regards to the role of ballmilling-induced microstructure for the thermal and mechanical impact in the Ni+Al system. First, the influence of the sequence and duration of applied milling/grinding steps on the microstructure of the produced composite particles was investigated. Second, the correlations between particles microstructures on both thermal and mechanical sensitivities of the reactive systems were revealed. Third, the kinetics and reaction mechanism for reactive medium with selected microstructures were experimentally explored. Fourth, the atomistic simulation of ignition processes was used to define the possible mechanism for the impact-induced reaction initiation. Finally, based on all of the above results, the nature of the relationship between thermal and mechanical initiation characteristics was established.

2. EXPERIMENTAL AND THEORETICAL METHODS 2.1. Materials Fabrication. Aluminum (Alfa Aesar, particle size 355 μm). Longer DM increases the portion of nanolaminated structure that was observed. For example, after tDM(8.5 min) + tWG(10 min), the fine nanostructure was observed in all size ranges (row 3). However, for larger particles, the coarse microstructure is still prevalent. Further increase in DM time results in the nanolaminated structure being dominant for almost all fractions (rows 4 and 5). Comparison of XRD patterns (Figure 7) for the initial Ni+Al mixture and F2 fractions of material obtained at tDM(4 min) + tWG(10 min) (only coarse microstructure) and tDM(17 min) + tWG(10 min) (only nanolaminated microstructure) conditions suggests that the applied ball-milling process does not lead to the formation of any intermetallic phases. It is important to note that whereas diffraction peaks are broader for the ballmilled powders, their positions are not shifted from those for the initial mixture. This indicates that metallic solid-solutions are not formed. These results were confirmed by EDS measurements of elemental compositions in Ni and Al layers showing, with accuracy of 0.1 wt %, the absence of metal dissolution. 3.2. Reactivity of Materials. Thermal Initiation of Reaction. Examination of the thermal ignition characteristics as a function of the distinct milling-induced microstructures showed that Tig for materials obtained by solely DM (17 min), where particles with coarse structure are dominant, is 720 ± 25 K. Note that this temperature is well below the eutectic in the Ni−Al system (912 K). Furthermore, it was demonstrated that varying the DM time and adding a WG step enables additional tuning of the thermal ignition temperature. Results of these experiments (Figure 8) suggest that, for example, at tDM(0 min) + tWG(10 min), the Tig decreases from 700 to 550 K with decreasing of particle sizes in the range 106 to 25 μm. Ignition behaviors for materials obtained at tDM(8.5 min) + tWG(10 min) and tDM(13 min) + tWG(10 min) were found to be close to each other. For these conditions, Tig for F1 fractions is below 575 K. At tDM(17 min) + tWG(10 min), the composite particles for all size fractions were observed to be the most thermally sensitive materials evaluated.

Figure 4. SEM images of initial Ni+Al mixture (a) and composite particles obtained after 4 (b) and 17 (c) min of dry ball milling.

Figure 5. High-magnification images for material obtained at 17 mn of ball milling (a), coarse (b), and nanolaminated microstructures (c,d).

250 μm thick) of nanolaminated microstructure. Some small amount (